JP2006060335A - Image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus which executes a block noise reduction process having little error detection even in a moving image region with severe luminance change by an MPEG2 system in the signal of the MPEG2 system or a video image signal having an edge component. <P>SOLUTION: The image processing apparatus includes a block boundary detection means 2 for detecting a block boundary line by judging the presence of a block noise between pixel blocks of an inputted luminance signal, a pixel block average luminance detection means 4 for calculating and detecting the average luminance for every pixel block unit of the inputted luminance signal, and a filter circuit 1 for performing the filtering process of the block noise based on a block boundary detection result and the average luminance detection result for every pixel block. When inputted compression image signal is displayed, the process of filtering is changed by the correlation of the front and rear pixel block of the block boundary line and the front and rear pixel block before one block, and the block noise is alleviated. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention is a television receiver (hereinafter referred to as “TV”) having a liquid crystal, PDP (plasma display), PALC (plasma addressed liquid crystal), organic EL, CRT (cathode tube) or the like as a display device, or a monitor with a DVD or the like. When displaying an MPEG2 system signal or displaying an MPEG2 system broadcast such as terrestrial digital broadcasting, processing of the system is detected, and block noise peculiar to the MPEG2 system is detected according to information obtained from the video signal. In particular, the present invention relates to an image processing apparatus capable of performing and reducing the optimum filtering process.

  In recent years, flooding of MPEG2 format signal sources has been progressing in DVD, digital broadcasting, and the like. Since the MPEG2 compression method is irreversible, video loss due to compression is inevitable. Examples of missing images include missing high-frequency information and mosquito noise, but the biggest problem is the occurrence of 8 × 8 block noise accompanying compression in units of pixel blocks. In some cases, such as when moving at high speed, when the screen is panned, or when there is a large amount of motion change, such as a scene with a large moving area such as the sea or waves, or where there is a small amount of change in the gradation, the rectangular image may be partially broken. It is done. Furthermore, video loss such as block noise occurs more frequently with a higher compression rate, that is, with a lower bit rate, which is a factor in image quality degradation.

  The MPEG2 system is mainly composed of three types of frames. It is a start point that is encoded independently of other frames, and is an I (Intra-picture) frame that is subjected to compression processing almost equivalent to JPEG and P ( There are a Predictive Picture (B) frame and a B (Bidirectionally Predictive Picture) frame obtained by interpolation from the preceding and following I and P frames. The basic processing block in these interpolation processing and motion compensation processing is a pixel block obtained by dividing the corresponding image into 8 × 8 pixels. In order to reduce the amount of information by removing high frequency components in the image, that is, details, DCT (Discrete Cosine Transform) is performed for each pixel block, quantization is performed, and motion vector detection is performed (patent) References 1 and 2).

Due to the features of these MPEG2 encoding processes, an error due to information compression occurs between pixel block units as block noise, which causes significant image quality degradation especially in a low bit rate video signal.
JP-A-10-276433 JP-A-9-275560

  There are roughly two methods for reducing block noise. The first is a method of changing the filter coefficient by acquiring bit rate information from the MPEG decoder and performing a high-intensity filtering process when the bit rate is low, or a low-intensity filtering process when the bit rate is high ( (See FIG. 3). Although this method can reduce block noise, it may use a fixed passive filter, which also removes video components other than block noise. Become.

  The other method is a method in which the block boundary line position is detected from the video signal and only the pixel block boundary line is filtered (see FIG. 4). This method can remove block noise without losing the detail of the entire video, but it detects the edge of a signal such as a vertical line as a block boundary line, which may result in edge blurring or change in luminance. For example, erroneous detection is performed with a severe signal, which causes deterioration of image quality. This is because the process detects and determines only pixel changes in the signal level of the screen, and thus cannot distinguish whether the edge is an image itself or an edge caused by block noise.

  Therefore, the present invention provides an image processing apparatus capable of performing block noise reduction processing that is less likely to be erroneously detected even in a moving image region having a sharp luminance change or a video signal having an edge component. The purpose is to do.

In order to solve the above problems, the present invention has the following configuration.
The video processing apparatus of the present invention includes a timing generating means for generating a timing pulse for detecting an 8 × 8 pixel block from a horizontal synchronizing signal, a vertical synchronizing signal, and a clock signal, and an 8 × 8 from a luminance signal inputted to the timing pulse. A pixel block boundary detecting unit that detects a pixel block boundary line and generates a pixel block boundary signal, a pixel block average luminance detecting unit that detects an average signal level of the pixel block from the timing pulse and the luminance signal, and a pixel one frame before Difference information storage means for storing the average signal level of the block, filter control means for determining the coefficient of the filter from the pixel block boundary signal and the average signal level signal, and filtering the input signal based on the filter control signal to output the output signal It has the structure which consists of the filter circuit which outputs.

  An image processing apparatus according to the present invention includes a block boundary detection unit that detects a block boundary line by determining the presence or absence of block noise between pixel blocks of an input luminance signal, and an average of the input luminance signal for each pixel block. A pixel block average luminance detecting means for calculating and detecting luminance, and a filter circuit that performs block noise filtering processing based on a block boundary detection result and an average luminance detection result for each pixel block, and an input compressed image When displaying a signal, the block noise is reduced by changing the filtering process according to the correlation between the pixel block before and after the block boundary line and the pixel block before and after the previous frame. .

  In the image processing apparatus of the present invention, when the block boundary detection means performs block boundary detection, the threshold value is changed according to the type of the signal source to determine the presence or absence of block noise.

  The image processing apparatus of the present invention has timing generation means for setting an arbitrary area of the screen as a detection area when generating a timing generation signal for separating the pixel block.

  The image processing apparatus of the present invention includes a filter control unit that stops the filtering process according to the type of the signal source.

  The image processing apparatus according to the present invention includes a CPU that controls the block boundary detection unit, the timing generation unit, or the filter control unit according to a control signal to stop the operation of the filtering process.

  In the image processing apparatus of the present invention, the CPU has setting values tabulated and performs different settings for each input signal.

  According to the present invention, it is possible to perform block noise reduction processing that does not cause false detection even in a moving image region where the luminance changes drastically or a video signal having an edge component.

The best mode for carrying out the present invention will be described below in detail with reference to the drawings.
An image processing apparatus according to an embodiment of the present invention will be described below with reference to FIG. The video processing apparatus of this embodiment includes a filter circuit 1, a pixel block boundary detection (means) 2, a timing generation (means) 3, a pixel block average luminance detection (means) 4, and a difference information storage (means). 5 and a filter control (means) 6.

  The horizontal synchronization signal Hsync, the vertical synchronization signal Vsync, and the clock signal Clock input from the input terminals Ih, Iv, and Iclk are input to the timing generation (means) 3. Timing generation (means) 3 generates a timing pulse for determining an 8 × 8 pixel block based on a clock signal Clock from a horizontal start point and a vertical start line arbitrarily determined from the horizontal synchronization signal Hsync and the vertical synchronization signal Vsync. The timing pulse a is transmitted to the block boundary detection (means) 2 and the pixel block difference detection (means) 4. FIG. 5 shows an example of implementation.

  The block boundary detection (means) 2 detects a video signal around the front and rear pixels specified by the timing pulse a from the luminance signal Y input from the input terminal Iy, and the difference between the front and rear pixels is equal to or greater than an arbitrary threshold value. (Detection A) and the presence of block noise between the corresponding pixel blocks by detecting that the pixels around the block boundary of the pixel block separated by the timing pulse a are below an arbitrary difference range (detection B) Make a decision. FIG. 6 shows an example of implementation. The determination result is transmitted to the filter control (means) 6 as a pixel block boundary signal b.

  On the other hand, the pixel block average luminance (difference) detection (means) 4 divides the input luminance signal Y for each pixel block by the timing pulse a, and calculates the average signal level. The calculation result is recorded in the average luminance information storage (means) 5.

  The filter control (means) 6 changes the filter setting according to the average luminance level of the current block at the boundary and the pixel blocks before and after the previous frame when the boundary is recognized by the pixel block boundary signal b.

  An example of the control method will be described based on FIG. When the difference in the average luminance level between the pixel block A and the pixel block C one frame before is more than a certain level, and the difference in the average luminance level between the pixel block B and the pixel block D one frame before is more than a certain level, It is thought that there was no correlation and a scene change occurred. When a scene change occurs, there is no correlation between the previous and next frames, and a lot of block noise occurs. Therefore, a filtering process using a strong filter as shown in FIG. 8 is performed.

  Further, the difference between the average luminance levels of the pixel block A and the pixel block C one frame before is less than a certain value, and the difference between the average luminance levels of the pixel block B and the pixel block D one frame before is less than a certain value, When the difference in the average luminance level between the pixel block A and the pixel block B is below a certain level, it can be considered as a still image region of a low gradation change portion. Since block noise in the still image area is relatively noticeable and causes image quality degradation, filtering processing using a medium intensity filter as shown in FIG. 8 is performed.

  Further, the difference between the average luminance levels of the pixel block A and the pixel block C one frame before is less than a certain value, and the difference between the average luminance levels of the pixel block B and the pixel block D one frame before is less than a certain value, If the difference between the average luminance levels of the pixel block A and the pixel block B is greater than or equal to a certain level, it is considered that there is edge information of a still image instead of block noise. If the filtering process is performed on such a video region, the detail of the entire video is lost, so the filtering process is not performed.

  Further, when the difference in the average luminance level between the pixel block A and the pixel block C one frame before is less than a certain level, and the difference in the average luminance level between the pixel block B and the pixel block D one frame before is equal to or more than a certain value, On the contrary, when the difference between the pixel blocks A and C is equal to or larger than the certain value and the difference between the pixel blocks B and D is equal to or smaller than the certain value, it is considered as a boundary region between the still region and the moving image region. Although this portion is also regarded as an edge, filtering processing by a weak intensity filter as shown in FIG. 8 is performed from the point that block noise occurs in the moving image region and the region where mosquito noise occurs.

  A summary of the above conditions is shown in FIG. Here, | A−C | >> α means that the difference in the average luminance level between the pixel block A and the pixel block C one frame before is equal to or greater than a certain value, and the other is the same.

1 is a block diagram showing a circuit configuration of an image processing apparatus according to Embodiment 1 of the present invention. FIG. 6 is a block diagram showing a circuit configuration of an image processing apparatus according to Embodiment 2 of the present invention. FIG. 1 is a block diagram showing a circuit configuration of a conventional technique. FIG. 2 is a block diagram showing a circuit configuration of a conventional technique. The operation | movement explanatory drawing of a timing generation circuit. The operation | movement explanatory drawing of a block boundary detection circuit. The correlation diagram of a pixel block. An example of filtering processing. List of filter control contents.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Filter circuit 2 Block boundary detection 3 Timing generation 4 Pixel block average brightness detection 5 Average brightness information storage 6 Filter control 7 CPU

Claims (6)

  Block boundary detection means for detecting a block boundary line by determining the presence or absence of block noise between pixel blocks of the input luminance signal, and a pixel that is detected by calculating an average luminance for each pixel block of the input luminance signal A block average luminance detection means, and a filter circuit that performs a block noise filtering process based on a block boundary detection result and an average luminance detection result for each pixel block, and displays a block image when an input compressed image signal is displayed. An image processing apparatus that reduces block noise by changing a filtering process according to a correlation between a pixel block before and after a boundary line and a pixel block before and after one frame.   The image processing apparatus according to claim 1, wherein the block boundary detection unit determines the presence or absence of block noise by changing a threshold value according to a type of a signal source when performing block boundary detection.   The image processing apparatus according to claim 1, further comprising timing generation means for setting an arbitrary area of the screen as a detection area when generating the timing generation signal for separating the pixel blocks.   The image processing apparatus according to claim 1, further comprising a filter control unit that stops the filtering process according to a type of the signal source.   The image processing according to any one of claims 1 to 4, wherein the block boundary detection unit, the timing generation unit, or the filter control unit includes a CPU that controls a threshold value and an operation stop of the filtering process by a control signal. apparatus.   The image processing apparatus according to claim 5, wherein the CPU has setting values tabulated and performs different settings for each input signal.

Images